- Email: [email protected]
TRANSFORMING GROWTH FACTOR β1 AND SOLUBLE FAS SERUM LEVELS IN HEPATOCELLULAR CARCINOMA
doi:10.1006/cyto.1999.0650, available online at http://www.idealibrary.com on
SHORT COMMUNICATION
TRANSFORMING GROWTH FACTOR 1 AND SOLUBLE FAS SERUM LEVELS IN HEPATOCELLULAR CARCINOMA Rodolfo Sacco, Domenico Leuci, Cosimo Tortorella, Giorgio Fiore, Felice Marinosci, Oronzo Schiraldi, Salvatore Antonaci In this study we assessed the usefulness of serum Transforming Growth Factor-1 (TGF-1) and soluble Fas (sFas) in distinguishing liver cirrhosis (LC) with and without hepatocellular carcinoma (HCC) as compared with alpha-fetoprotein (AFP). Serum TGF-1 and sFas levels were measured by ELISA in 51 LC patients, 54 patients with HCC and 30 healthy donors. Considering as a cut-off limit (mean+1SD of controls) 74 pg/ml and 637 pg/ml for TGF-1 and sFas, respectively, we computed serum concentrations of TGF-1 and sFas as a score (meanSD). The positive frequency of serum TGF-1 levels in HCC patients (54%) was greater than in LC patients (26%) and healthy donors (3%). TGF-1 levels were higher in HCC (1.60.5) than in LC (1.10.2) (P<0.0001) and healthy donors (0.60.2). Using a cut-off limit of 82 pg/ml (mean+2SD), the positive frequency of TGF-1 was 20% in HCC patients. None of the controls and LC patients had TGF-1 levels higher than 82 pg/ml. The positive frequency of serum sFas levels was 100% in HCC patients, 98% in LC patients and 3% in healthy controls. Serum sFas levels were higher in HCC (2.50.7) than in LC (1.90.5) (P<0.001) and healthy donors (0.60.3). No significant change of positive frequency was obtained by setting sFas cut-off at higher levels. sFas values did not correlate with TGF-1 levels. No relationship was found between TGF-1 amounts and AFP levels. However, in the 23% of HCC patients, with normal AFP values TGF-1 levels were higher than the cut off. These findings suggest the potential usefulness for TGF-1 assay in AFP-negative HCC. 2000 Academic Press
Alpha-fetoprotein (AFP) is the most frequently used marker in the screening for hepatocellular carcinoma (HCC). However, up to 30% of HCC have normal AFP levels.1,2 Several observations indicate that the production of growth factors by various cells may be involved in neoplastic transformation and tumour development.3 Transforming Growth Factor-1 (TGF-1) is involved in cellular matrix formation and inhibition of both normal and neoplastic hepatocyte proliferation.4 From the Department of Internal Medicine, Immunology and Infectious Diseases, University of Bari Medical School, Policlinico 70124 Bari, Italy Correspondence to: Rodolfo Sacco, Department of Internal Medicine, Immunology and Infectious Diseases, c/o Clinica Medica II, Policlinico, 70124 Bari, Italy; E-mail: [email protected] Received 22 February 1999; received in revised form 6 September 1999; accepted for publication 20 September 1999 2000 Academic Press 1043–4666/00/060811+04 $35.00/0 KEY WORDS: cirrhosis/HCC/soluble Fas/Transforming Growth Factor 1/tumour markers CYTOKINE, Vol. 12, No. 6 (June), 2000: pp 811–814
Much attention has been recently focused on the role of Fas antigen (CD95) in liver cell damage during the course of viral hepatitis.5. Fas is also released in a soluble form (sFas), a phenomenon which provides a mechanism for tumour cells to escape immunosurveillance, thus favouring carcinogenesis.6 We determined serum levels of TGF-1 and sFas in patients with LC or LC-associated HCC to evaluate whether TGF-1 and sFas assay may provide a useful tool for early HCC diagnosis in clinical practice.
RESULTS The standard curve showed satisfactory linearity within the range of both TGF-1 and sFas concentration from 16 to 1000 pg/ml. In normal controls, serum TGF-1 and sFas levels were 668 pg/ml and 533104 pg/ml, respectively. Considering as a cut-off limit (mean+1SD of controls) 74 pg/ml and 637 pg/ml 811
812 / Sacco et al.
CYTOKINE, Vol. 12, No. 6 (June, 2000: 811–814)
3.5
5.5 5
3
4.5 4
2.5 sFas
TGF-β1
3.5 2 1.5
3 2.5 2
1
1.5 1
0.5
0.5 0
Controls n = 30
LC n = 51
0
HCC n = 54
Controls n = 30
LC n = 51
HCC n = 54
Figure 1. TGF-1 serum levels in LC patients, HCC subjects and healthy donors.
Figure 2. sFas serum levels in LC patients, HCC individuals and healthy subjects.
Values are computed as score. Horizontal bars and dotted line represent median values and the cut-off value of TGF-1, respectively.
Values are computed as score. Horizontal bars and dotted line represent median values and the cut-off value of sFas, respectively.
10 r = 0.263 P = NS
sFas
for TGF-1 and sFas, respectively, we computed serum TGF-1 and sFas values as score (meanSD) in healthy donors and patient groups. The positive frequency of serum TGF-1 levels in HCC patients (54%) was greater than that in healthy donors (3%) and LC patients (26%) (P<0.0001 and P<0.01, respectively). In healthy controls serum TGF-1 levels were 0.60.2, while in patient groups high serum levels of TGF-1 (1.10.2 in subjects with LC and 1.60.5 in patients with HCC, respectively) were found (Fig. 1). The values of TGF-1 in both groups of subjects with liver disease were higher than those seen in controls (P<0.0001 for both LC and HCC). TGF-1 levels were higher in HCC than in LC (P<0.0001). When the cut-off limit was set at 82 pg/ml (mean+2SD), the positive frequency of TGF-1 was 20% in HCC patients. None of the controls and LC patients had TGF-1 levels higher than 82 pg/ml. The positive frequency of serum sFas levels was 100% in HCC patients, 98% in LC patients and 3% in healthy controls. This gave rise to a statistical significance (P<0.0001) in comparison to controls for both patient groups, while there were no differences between LC and HCC individuals. Also sFas serum values in patient groups were higher than in controls, as follows: 1.90.5 in patients with LC, 2.50.7 in patients with HCC and 0.60.3 in healthy controls (P<0.0001 for both LC and HCC vs controls (Fig. 2). Moreover, serum sFas levels were significantly increased (P<0.001) in patients with HCC compared with LC. However, no significant differences in terms of positive frequency were observed between patient groups by setting sFas cut-off level at 741 pg/ml (mean+2SD). Serum AFP concentrations were assessed in the two groups of patients and expressed as scoreSD,
1
1 TGF-β1
0.1 Figure 3.
10
Relationship between sFas and TGF-1 score.
Vertical and horizontal dotted lines represent cut-off value of sFas and TGF-1, respectively.
considering as cut-off limits 7 IU/ml. AFP levels were found to be 5.59 in 15 of 51 (30%) LC patients and 61.8126.8 in 41 of 54 (76%) HCC subjects (P<0.0001 and P<0.01 for positive frequency and levels, respectively, in HCC vs LC patients). Serum sFas values were not correlated to TGF-1 levels (Fig. 3). This was also the case when serum TGF-1 and AFP levels were compared (Fig. 4). However, three of 13 (23%) HCC patients, whose AFP values fell within normal range, had TGF-1 levels higher than the cut-off 1.70.8).
DISCUSSION Growing attention has recently been given to the possible role of impaired programmed cell death or
Serum TGF1 and sFas in HCC / 813
10
TGF-β1
r = 0.241 P = NS
1
0.1
1
10
100
1000
AFP Figure 4.
Furthermore, no correlation between serum sFas levels and serum TGF-1 was detected. These findings indicate that, even if sFas is closely associated with HCC, its serum determination might be of poor utility for detection of HCC in patients with LC. In conclusion, our results show that serum TGF-1 and sFas levels are significantly higher in patients with HCC than patients with LC and normal subjects. However in the light of our findings, the use of serum TGF-1 levels assay may only significantly improve the detection of AFP-negative HCC in high-risk patients.
Relationship between TGF1 and AFP in HCC patients.
Values are computed as score. Vertical and horizontal dotted lines represent cut-off value of TGF-1 and AFP.
apoptosis in carcinogenesis. Whereas in some cases apoptosis is strictly related to growth factor failure, in others it is actively induced by positive signals such as those triggered by TGF-1 and Fas system.7 Our study points out that LC and HCC patients have serum TGF-1 levels significantly higher than normal subjects. Patients with HCC also showed increased values of serum TGF-1 compared with LC patients. Although the underlying liver disease may per se account for the elevation of serum TGF-1 levels,8 our data indicate that the amount of serum TGF-1 in patients with HCC may be in part related to production by the HCC cells themselves. The observation of a TGF-1 gene overexpression in human HCC cell lines further supports this possibility.9 Our findings suggest that serum TGF-1 may be a useful marker for HCC. We found that the positive frequency of serum TGF-1 levels is significantly higher in HCC patients than LC patients and/or healthy controls. Moreover, by setting the cut-off level at mean+2SD, the positive frequency of TGF-1 was 20% in HCC patients, whereas none of the cirrhotic patients exhibited higher levels than this cut-off, implying specificity of serum TGF-1 as HCC-associated marker. Furthermore, although no correlation between serum TGF-1 and AFP was found, in the 23% of HCC patients showing AFP value within normal limits, TGF-1 levels are higher than cut-off. Therefore, even though the simultaneous determination of both serum TGF-1 and AFP levels may improve the detection of HCC, our data indicate that the evaluation of serum TGF-1 may be a useful tool above all for the diagnosis of AFP-negative HCC in high-risk groups. Our results also showed a significant elevation of serum sFas levels in patients with HCC compared with LC and healthy controls. Despite the difference in serum sFas levels between HCC and LC patients, the positive frequency of serum sFas was similar in the two groups, even when setting the cut-off at mean+2SD.
MATERIALS AND METHODS Fifty-one patients with LC (31 males and 20 females, aged 22–74 years; mean age 5410) and 54 patients with LC-associated HCC (34 males and 20 females, aged 51–84 years; mean age 677) admitted to the Department of Internal Medicine, University of Bari, Italy, from May 1997 to July 1998 were consecutively enrolled for this study. None of the patients with HCC underwent treatment before collecting serum samples. All patients underwent serological evaluation for hepatitis B virus (HBV), hepatitis D virus (HDV) or hepatitis C virus (HCV) infections. HBV and HDV markers were tested using commercially available ELISA kits for HBsAg, anti-HBc, anti-HDV (SORIN Biomedica, Saluggia, Italy). The third generation ELISA assay (Ortho Diagnostics, Raritan, NJ) was used to detect serum anti-HCV antibodies. Intravenous drug addicts, male homosexuals and patients with liver disease related to alcohol and hepatotoxic drugs, to autoimmune or inherited metabolic disorders were excluded from the study. LC was diagnosed by clinical features, laboratory tests, ultrasonography (US), computed tomography (CT), and was confirmed by histology. Diagnosis of HCC was performed histologically and/or by laboratory tests, US, CT, magnetic resonance imaging and selective hepatic arteriography. As control, 30 healthy adults, 16 males and 14 females (mean age 508 years, range 29–52) were also included. They were negative for HBsAg, anti-HDV antibodies, anti-HCV antibodies and showed normal liver function. Serum TGF-1 and sFas levels were measured using the TGF-1 and sFas ELISA kits (Amersham, Buckinghamshire, UK and Bender Medsystem, Wien, Austria, respectively). The sensitivity of the assay was 4 pg/ml for TGF-1 and 16 pg/ml for sFas. The concentration of TGF-1 and sFas in the samples was determined by interpolation from a standard curve according to manufacturer’s instructions. Patients’ sera were also tested for AFP using AFP-ELISA kit (Abbott Laboratories North Chicago, IL). The level differences between the means of continuous variables were compared with unpaired and paired Student’s t-test. Spearman’s correlation test in which correlation coefficients were computed by linear least squares regression was performed . P<0.05 was considered statistically significant.
814 / Sacco et al.
REFERENCES 1. Pateron D, Ganne N, Trinchet JC, Aurousseau MH, Mal F, Meicler C, Coderc E, Reboullet P, Beaugrand M (1994) Prospective study of screening for hepatocellular carcinoma in Caucasian patients with cirrhosis. J Hepatol 20:65–71. 2. Tsai JF, Chang WY, Jeng JE, Ho MS, Lin ZY, Tsai JH (1994) Frequency of raised AFP level among Chinese patients with hepatocellular carcinoma related to hepatitis B and C. Br J Cancer 69:1157–1159. 3. Cross MC, Dexter TM (1991) Growth factors in development transformation and tumorigenesis. Cell 64:271–280. 4. Fausto N, Mead JE, Gruppuso PA, Brown L (1991) TGFbeta in liver development, regeneration and carcinogenesis. Ann NY Acad Sci 593:231–242.
CYTOKINE, Vol. 12, No. 6 (June, 2000: 811–814) 5. Galle PR, Hofmann WJ, Walczak H, Schaller H, Otto G, Stremmel W, Krammer PH, Runkel L (1995) Involvement of CD95 (APO-1/Fas) receptor and ligand in liver damage. J Exp Med 182:1223–1230. 6. Cheng J, Zhou T, Liu C, Shapiro JP, Brauer MJ, Kiafer MC, Barr PJ, Mountz JD (1994) Protection from Fas-mediated apoptosis by a soluble form of the Fas molecule. Science 263: 1759–1762. 7. Thompson CB (1995) Apoptosis in the pathogenesis and treatment of disease. Science 267:1456–1462. 8. Castilla A, Prieto J, Fausto N (1991) The transforming growth factor 1 and in chronic liver disease: effect of interferon alpha therapy. N Engl J Med 323:933–940. 9. Ito N, Kawata S, Tamura S, Takaishi K, Yabuuchi Y, Matsuda Y, Nishiora M, Tarui S (1990) Expression of Transforming growth factor 1 mRNA in human hepatocellular carcinoma. Jpn J Cancer Res 81:1202–1205.
SHORT COMMUNICATION
TRANSFORMING GROWTH FACTOR 1 AND SOLUBLE FAS SERUM LEVELS IN HEPATOCELLULAR CARCINOMA Rodolfo Sacco, Domenico Leuci, Cosimo Tortorella, Giorgio Fiore, Felice Marinosci, Oronzo Schiraldi, Salvatore Antonaci In this study we assessed the usefulness of serum Transforming Growth Factor-1 (TGF-1) and soluble Fas (sFas) in distinguishing liver cirrhosis (LC) with and without hepatocellular carcinoma (HCC) as compared with alpha-fetoprotein (AFP). Serum TGF-1 and sFas levels were measured by ELISA in 51 LC patients, 54 patients with HCC and 30 healthy donors. Considering as a cut-off limit (mean+1SD of controls) 74 pg/ml and 637 pg/ml for TGF-1 and sFas, respectively, we computed serum concentrations of TGF-1 and sFas as a score (meanSD). The positive frequency of serum TGF-1 levels in HCC patients (54%) was greater than in LC patients (26%) and healthy donors (3%). TGF-1 levels were higher in HCC (1.60.5) than in LC (1.10.2) (P<0.0001) and healthy donors (0.60.2). Using a cut-off limit of 82 pg/ml (mean+2SD), the positive frequency of TGF-1 was 20% in HCC patients. None of the controls and LC patients had TGF-1 levels higher than 82 pg/ml. The positive frequency of serum sFas levels was 100% in HCC patients, 98% in LC patients and 3% in healthy controls. Serum sFas levels were higher in HCC (2.50.7) than in LC (1.90.5) (P<0.001) and healthy donors (0.60.3). No significant change of positive frequency was obtained by setting sFas cut-off at higher levels. sFas values did not correlate with TGF-1 levels. No relationship was found between TGF-1 amounts and AFP levels. However, in the 23% of HCC patients, with normal AFP values TGF-1 levels were higher than the cut off. These findings suggest the potential usefulness for TGF-1 assay in AFP-negative HCC. 2000 Academic Press
Alpha-fetoprotein (AFP) is the most frequently used marker in the screening for hepatocellular carcinoma (HCC). However, up to 30% of HCC have normal AFP levels.1,2 Several observations indicate that the production of growth factors by various cells may be involved in neoplastic transformation and tumour development.3 Transforming Growth Factor-1 (TGF-1) is involved in cellular matrix formation and inhibition of both normal and neoplastic hepatocyte proliferation.4 From the Department of Internal Medicine, Immunology and Infectious Diseases, University of Bari Medical School, Policlinico 70124 Bari, Italy Correspondence to: Rodolfo Sacco, Department of Internal Medicine, Immunology and Infectious Diseases, c/o Clinica Medica II, Policlinico, 70124 Bari, Italy; E-mail: [email protected] Received 22 February 1999; received in revised form 6 September 1999; accepted for publication 20 September 1999 2000 Academic Press 1043–4666/00/060811+04 $35.00/0 KEY WORDS: cirrhosis/HCC/soluble Fas/Transforming Growth Factor 1/tumour markers CYTOKINE, Vol. 12, No. 6 (June), 2000: pp 811–814
Much attention has been recently focused on the role of Fas antigen (CD95) in liver cell damage during the course of viral hepatitis.5. Fas is also released in a soluble form (sFas), a phenomenon which provides a mechanism for tumour cells to escape immunosurveillance, thus favouring carcinogenesis.6 We determined serum levels of TGF-1 and sFas in patients with LC or LC-associated HCC to evaluate whether TGF-1 and sFas assay may provide a useful tool for early HCC diagnosis in clinical practice.
RESULTS The standard curve showed satisfactory linearity within the range of both TGF-1 and sFas concentration from 16 to 1000 pg/ml. In normal controls, serum TGF-1 and sFas levels were 668 pg/ml and 533104 pg/ml, respectively. Considering as a cut-off limit (mean+1SD of controls) 74 pg/ml and 637 pg/ml 811
812 / Sacco et al.
CYTOKINE, Vol. 12, No. 6 (June, 2000: 811–814)
3.5
5.5 5
3
4.5 4
2.5 sFas
TGF-β1
3.5 2 1.5
3 2.5 2
1
1.5 1
0.5
0.5 0
Controls n = 30
LC n = 51
0
HCC n = 54
Controls n = 30
LC n = 51
HCC n = 54
Figure 1. TGF-1 serum levels in LC patients, HCC subjects and healthy donors.
Figure 2. sFas serum levels in LC patients, HCC individuals and healthy subjects.
Values are computed as score. Horizontal bars and dotted line represent median values and the cut-off value of TGF-1, respectively.
Values are computed as score. Horizontal bars and dotted line represent median values and the cut-off value of sFas, respectively.
10 r = 0.263 P = NS
sFas
for TGF-1 and sFas, respectively, we computed serum TGF-1 and sFas values as score (meanSD) in healthy donors and patient groups. The positive frequency of serum TGF-1 levels in HCC patients (54%) was greater than that in healthy donors (3%) and LC patients (26%) (P<0.0001 and P<0.01, respectively). In healthy controls serum TGF-1 levels were 0.60.2, while in patient groups high serum levels of TGF-1 (1.10.2 in subjects with LC and 1.60.5 in patients with HCC, respectively) were found (Fig. 1). The values of TGF-1 in both groups of subjects with liver disease were higher than those seen in controls (P<0.0001 for both LC and HCC). TGF-1 levels were higher in HCC than in LC (P<0.0001). When the cut-off limit was set at 82 pg/ml (mean+2SD), the positive frequency of TGF-1 was 20% in HCC patients. None of the controls and LC patients had TGF-1 levels higher than 82 pg/ml. The positive frequency of serum sFas levels was 100% in HCC patients, 98% in LC patients and 3% in healthy controls. This gave rise to a statistical significance (P<0.0001) in comparison to controls for both patient groups, while there were no differences between LC and HCC individuals. Also sFas serum values in patient groups were higher than in controls, as follows: 1.90.5 in patients with LC, 2.50.7 in patients with HCC and 0.60.3 in healthy controls (P<0.0001 for both LC and HCC vs controls (Fig. 2). Moreover, serum sFas levels were significantly increased (P<0.001) in patients with HCC compared with LC. However, no significant differences in terms of positive frequency were observed between patient groups by setting sFas cut-off level at 741 pg/ml (mean+2SD). Serum AFP concentrations were assessed in the two groups of patients and expressed as scoreSD,
1
1 TGF-β1
0.1 Figure 3.
10
Relationship between sFas and TGF-1 score.
Vertical and horizontal dotted lines represent cut-off value of sFas and TGF-1, respectively.
considering as cut-off limits 7 IU/ml. AFP levels were found to be 5.59 in 15 of 51 (30%) LC patients and 61.8126.8 in 41 of 54 (76%) HCC subjects (P<0.0001 and P<0.01 for positive frequency and levels, respectively, in HCC vs LC patients). Serum sFas values were not correlated to TGF-1 levels (Fig. 3). This was also the case when serum TGF-1 and AFP levels were compared (Fig. 4). However, three of 13 (23%) HCC patients, whose AFP values fell within normal range, had TGF-1 levels higher than the cut-off 1.70.8).
DISCUSSION Growing attention has recently been given to the possible role of impaired programmed cell death or
Serum TGF1 and sFas in HCC / 813
10
TGF-β1
r = 0.241 P = NS
1
0.1
1
10
100
1000
AFP Figure 4.
Furthermore, no correlation between serum sFas levels and serum TGF-1 was detected. These findings indicate that, even if sFas is closely associated with HCC, its serum determination might be of poor utility for detection of HCC in patients with LC. In conclusion, our results show that serum TGF-1 and sFas levels are significantly higher in patients with HCC than patients with LC and normal subjects. However in the light of our findings, the use of serum TGF-1 levels assay may only significantly improve the detection of AFP-negative HCC in high-risk patients.
Relationship between TGF1 and AFP in HCC patients.
Values are computed as score. Vertical and horizontal dotted lines represent cut-off value of TGF-1 and AFP.
apoptosis in carcinogenesis. Whereas in some cases apoptosis is strictly related to growth factor failure, in others it is actively induced by positive signals such as those triggered by TGF-1 and Fas system.7 Our study points out that LC and HCC patients have serum TGF-1 levels significantly higher than normal subjects. Patients with HCC also showed increased values of serum TGF-1 compared with LC patients. Although the underlying liver disease may per se account for the elevation of serum TGF-1 levels,8 our data indicate that the amount of serum TGF-1 in patients with HCC may be in part related to production by the HCC cells themselves. The observation of a TGF-1 gene overexpression in human HCC cell lines further supports this possibility.9 Our findings suggest that serum TGF-1 may be a useful marker for HCC. We found that the positive frequency of serum TGF-1 levels is significantly higher in HCC patients than LC patients and/or healthy controls. Moreover, by setting the cut-off level at mean+2SD, the positive frequency of TGF-1 was 20% in HCC patients, whereas none of the cirrhotic patients exhibited higher levels than this cut-off, implying specificity of serum TGF-1 as HCC-associated marker. Furthermore, although no correlation between serum TGF-1 and AFP was found, in the 23% of HCC patients showing AFP value within normal limits, TGF-1 levels are higher than cut-off. Therefore, even though the simultaneous determination of both serum TGF-1 and AFP levels may improve the detection of HCC, our data indicate that the evaluation of serum TGF-1 may be a useful tool above all for the diagnosis of AFP-negative HCC in high-risk groups. Our results also showed a significant elevation of serum sFas levels in patients with HCC compared with LC and healthy controls. Despite the difference in serum sFas levels between HCC and LC patients, the positive frequency of serum sFas was similar in the two groups, even when setting the cut-off at mean+2SD.
MATERIALS AND METHODS Fifty-one patients with LC (31 males and 20 females, aged 22–74 years; mean age 5410) and 54 patients with LC-associated HCC (34 males and 20 females, aged 51–84 years; mean age 677) admitted to the Department of Internal Medicine, University of Bari, Italy, from May 1997 to July 1998 were consecutively enrolled for this study. None of the patients with HCC underwent treatment before collecting serum samples. All patients underwent serological evaluation for hepatitis B virus (HBV), hepatitis D virus (HDV) or hepatitis C virus (HCV) infections. HBV and HDV markers were tested using commercially available ELISA kits for HBsAg, anti-HBc, anti-HDV (SORIN Biomedica, Saluggia, Italy). The third generation ELISA assay (Ortho Diagnostics, Raritan, NJ) was used to detect serum anti-HCV antibodies. Intravenous drug addicts, male homosexuals and patients with liver disease related to alcohol and hepatotoxic drugs, to autoimmune or inherited metabolic disorders were excluded from the study. LC was diagnosed by clinical features, laboratory tests, ultrasonography (US), computed tomography (CT), and was confirmed by histology. Diagnosis of HCC was performed histologically and/or by laboratory tests, US, CT, magnetic resonance imaging and selective hepatic arteriography. As control, 30 healthy adults, 16 males and 14 females (mean age 508 years, range 29–52) were also included. They were negative for HBsAg, anti-HDV antibodies, anti-HCV antibodies and showed normal liver function. Serum TGF-1 and sFas levels were measured using the TGF-1 and sFas ELISA kits (Amersham, Buckinghamshire, UK and Bender Medsystem, Wien, Austria, respectively). The sensitivity of the assay was 4 pg/ml for TGF-1 and 16 pg/ml for sFas. The concentration of TGF-1 and sFas in the samples was determined by interpolation from a standard curve according to manufacturer’s instructions. Patients’ sera were also tested for AFP using AFP-ELISA kit (Abbott Laboratories North Chicago, IL). The level differences between the means of continuous variables were compared with unpaired and paired Student’s t-test. Spearman’s correlation test in which correlation coefficients were computed by linear least squares regression was performed . P<0.05 was considered statistically significant.
814 / Sacco et al.
REFERENCES 1. Pateron D, Ganne N, Trinchet JC, Aurousseau MH, Mal F, Meicler C, Coderc E, Reboullet P, Beaugrand M (1994) Prospective study of screening for hepatocellular carcinoma in Caucasian patients with cirrhosis. J Hepatol 20:65–71. 2. Tsai JF, Chang WY, Jeng JE, Ho MS, Lin ZY, Tsai JH (1994) Frequency of raised AFP level among Chinese patients with hepatocellular carcinoma related to hepatitis B and C. Br J Cancer 69:1157–1159. 3. Cross MC, Dexter TM (1991) Growth factors in development transformation and tumorigenesis. Cell 64:271–280. 4. Fausto N, Mead JE, Gruppuso PA, Brown L (1991) TGFbeta in liver development, regeneration and carcinogenesis. Ann NY Acad Sci 593:231–242.
CYTOKINE, Vol. 12, No. 6 (June, 2000: 811–814) 5. Galle PR, Hofmann WJ, Walczak H, Schaller H, Otto G, Stremmel W, Krammer PH, Runkel L (1995) Involvement of CD95 (APO-1/Fas) receptor and ligand in liver damage. J Exp Med 182:1223–1230. 6. Cheng J, Zhou T, Liu C, Shapiro JP, Brauer MJ, Kiafer MC, Barr PJ, Mountz JD (1994) Protection from Fas-mediated apoptosis by a soluble form of the Fas molecule. Science 263: 1759–1762. 7. Thompson CB (1995) Apoptosis in the pathogenesis and treatment of disease. Science 267:1456–1462. 8. Castilla A, Prieto J, Fausto N (1991) The transforming growth factor 1 and in chronic liver disease: effect of interferon alpha therapy. N Engl J Med 323:933–940. 9. Ito N, Kawata S, Tamura S, Takaishi K, Yabuuchi Y, Matsuda Y, Nishiora M, Tarui S (1990) Expression of Transforming growth factor 1 mRNA in human hepatocellular carcinoma. Jpn J Cancer Res 81:1202–1205.